Analysing Five Years of GRACE Equivalent Water Height Variations Using the Principal Component Analysis
In this study we use 59 monthly solutions (April 2002–May 2007) of the Earth’s gravity field obtained from the Gravity Recovery and Climate Experiment (GRACE) satellite mission, expressed in equivalent water height (EWH). The GRACE-derived EWH values are
- PDF / 1,386,836 Bytes
- 9 Pages / 547.087 x 737.008 pts Page_size
- 86 Downloads / 214 Views
Analysing Five Years of GRACE Equivalent Water Height Variations Using the Principal Component Analysis I.M. Anjasmara and M. Kuhn
Abstract In this study we use 59 monthly solutions (April 2002–May 2007) of the Earth’s gravity field obtained from the Gravity Recovery and Climate Experiment (GRACE) satellite mission, expressed in equivalent water height (EWH). The GRACEderived EWH values are analysed first in terms of the secular trend and RMS-variability before applying the statistically-based Principal Component Analysis (PCA), in order to obtain the most dominant spatial and temporal variations. On a global scale, we show that only 5 modes can express more than 80% of the total variability, including all major hydrological, cryospheric and post-glacial rebound signals. As expected, globally, the most dominant temporal variation is an annual signal followed by a secular trend. Apart from these well-known signals, we show that the PCA is able to reveal other periodic and a-periodic signals.
73.1 Introduction Since its launch on 17 March 2002, the GRACE twin satellite mission provides valuable information on continental water storage variability (e.g., Ramillien et al., 2004; Tapley et al., 2004a). Today, GRACE time-variable gravity observations are routinely provided with a monthly temporal resolution and spatial
I.M. Anjasmara () The Institute of Geoscience Research, Western Australian Centre for Geodesy, Curtin University of Technology, Perth, WA 6845, Australia e-mail: [email protected]
resolutions of approximately 200–500 km (e.g., Wahr et al., 1998; Nerem et al., 2003; Ramillien et al., 2004; Tapley et al., 2004b), allowing mass changes on both regional and global scales to be inferred. Up until today, many studies based on GRACE observations have studied the annual and secular mass variations producing the strongest time-variable signal in the Earth’s gravity field (e.g., Ramilien et al., 2004; Wahr et al., 2004; Anderson and Hinderer, 2005; Rowlands et al., 2005; Hinderer et al., 2006; Schmidt et al., 2006). Most of these studies use analysing techniques that require some prior information on the signals to be studied, such as the assumption of a periodic signal (e.g., annual or semi-annual), thus do not allow for the detection of other a-periodic signals. Only few studies so far involving GRACE data have used the statistically-based Principal Component Analysis (PCA) (e.g., Preisendorfer, 1988; Jolliffe, 2002), also known as EOF analysis, that does not require any prior information. With an increased number of monthly solutions available (e.g., few years), the application of the PCA becomes an alternative analysing tool that is already well established in other areas, such as meteorology and oceanography. De Viron et al. (2006) applied the PCA on GRACE data in order to detect inter-annual signals of the continental hydrology related to the El-Niño Southern Oscillation (ENSO) Index. Rangelova et al. (2007) have applied the PCA to GRACE data over North America in order to find correlations between Ea
Data Loading...
